Position Specific Interaction Dependent Scoring Technique for Virtual Screening Based on Weighted Protein−Ligand Interaction Fingerprint Profiles

Citations Over TimeTop 10% of 2009 papers

Abstract

The desire to exploit structural information to aid structure based design and virtual screening led to the development of the interaction fingerprint for analyzing, mining, and filtering the binding patterns underlying the complex 3D data. In this paper we introduce a new approach, weighted SIFt (or w-SIFt), extending the concept of SIFt to capture the relative importance of different binding interactions. The methodology presented here for determining the weights in w-SIFt involves utilizing a dimensionality reduction technique for eliminating linear redundancies in the data followed by a stochastic optimization. We find that the relative weights of the fingerprint bits provide insight into what interactions are critical in determining inhibitor potency. Moreover, the weighted interaction fingerprint can serve as an interpretable position dependent scoring function for ligand protein interactions.

Related Papers

• Survey of Feature Points Detection and Matching using SURF, SIFT and PCA-SIFT(2014)
• → Improvements of Local Descriptor in HOG/SIFT by BOF Approach(2014)5 cited
• → Implementation of Image Matching Algorithm Based on SIFT Features(2014)3 cited
• Application of multiple-layer SIFT to semantic concept detection(2011)
• → Performance Improvement of SIFT-based Copy-move Forgery Detection Using CSLBP Descriptor(2020)